Briyan | Building in Public

built a CLI that sets up production-ready Express + TypeScript in 60 seconds. no config hell. no boilerplate. no wasted hours. just npx zuro-cli init and you're running. building in public from 36 followers. let's see where this goes.

RAG > fine-tuning for most real-world apps. Your data changes. Models don’t. Update a document → seconds Retrain a model → weeks RAG is a live cheat sheet. Fine-tuning is a tattoo.

@KaiXCreator Outrage 🙂‍↕️

I am a Vibe Coder, scare me with one word

Interviewer: Why does the AI give confident wrong answers? because it's not looking anything up. it's doing one thing: guessing the next word that sounds right. it has no idea it's wrong. it just... continues.

your retrieval is 80% of your RAG quality. the LLM is just reading what you hand it. stop optimizing the model. fix the retrieval.

the best setups use both. hybrid search: dense vectors + sparse keywords. because meaning AND exact match together is better than picking a side.

Interviewer: Why does everyone jump straight to vector databases for RAG? because "semantic search" sounds smarter than ctrl+F. but here's what nobody tells you:

Interviewer: What is RAG? your AI doesn't know everything. so before it answers, you hand it the right pages. like an open-book exam. the AI reads your notes, then answers. without it? it's guessing from memory. and its memory has a cutoff date.

@BatsouElef zinlog.devbybriyan.com 2 min daily log → AI writes your full performance review automatically. lmk if you have questions 🙂

Time to promote your product. 🚀 Share that product URL!!

@asmah2107 engineers optimize for the thing they can debate in a meeting. nobody makes a slide deck about circuit breakers.

Spent 3 months debating which LLM to use for our agent. Spent 3 days dealing with the outage caused by having no circuit breaker. The model was never the risk.

2 minutes a day. AI writes your full performance review automatically. zinlog.devbybriyan.com

your performance review has nothing to do with how hard you worked. it's 100% about how well you documented what you did. the engineer who logged their wins got promoted. the one who worked 60-hour weeks and remembered nothing got "meets expectations." i built something to fix this. RT if your last review didn't reflect what you actually did 👇

your manager doesn't remember what you did last month. neither do you. that's why good people get passed over at review time. not because they didn't work hard. because they couldn't prove it. 2 minutes every evening fixes this.

@mannupaaji Next project review kab krna he paji?

Petition to rename useEffect() to doNotUseEffect()

building something for this

Hard work doesn’t get you promoted. Documented work does.

@Akintola_steve I will give you 1% cut from my monetization 😅

@briyan_dev Pay me money 🙂‍↔️

Damn! I just used a new font and it made my application look like paradise! The name starts with R and ends with Y.

@abhidoes Me to my self

crossed 500 followers😭❤️